Head-to-head comparison of fecal occult blood tests for colorectal neoplasia detection in Chinese outpatients: a multicenter diagnostic trial.

The optimal fecal occult blood test (FOBT) and hemoglobin (Hb) threshold for balancing diagnostic accuracy and endoscopic demand in colorectal cancer (CRC) screening among Chinese outpatients remained unclear. A prospective multicenter trial was conducted from January 2017 to April 2021 across eight tertiary hospitals. Eligible outpatients were enrolled, whose stool samples were analyzed using quantitative fecal immunochemical test (FIT), self-administered qualitative FIT, general qualitative FIT, and gFOBT. All participants underwent colonoscopy. The primary outcome was the sensitivity of FOBTs for CRC detection. A total of 2,930 participants were enrolled, with 2,618 participants meeting inclusion criteria and completing FOBTs and colonoscopies. At the threshold of 5.2 µg/g, quantitative FIT demonstrated comparable sensitivity for CRC detection (90.1%) to self-administered qualitative FIT (87.1%), general qualitative FIT (91.1%), and gFOBT (82.2%) (all P > 0.05). The positive rate of quantitative FIT (12.4%) was significantly lower than that of self-administered qualitative FIT (27.5%), general qualitative FIT (24.2%), and gFOBT (23.0%) (all P < 0.05). Quantitative FIT showed superior specificity (93.9%) compared to self-administered qualitative FIT (79.5%), general qualitative FIT (83.6%), and gFOBT (80.7%) (all P < 0.05). Furthermore, quantitative FIT at 5.2 µg/g exhibited higher PPV and LR + for CRC, advanced adenoma (AA) and advanced neoplasia (AN) than other FOBTs. Receiver operating characteristic (ROC) analysis and area under the curve (AUC) revealed excellent accuracy for CRC detection (AUC: 0.951, 95% CI: 0.920-0.981). The quantitative FIT with a threshold of 5.2 µg/g demonstrated superior performance for early CRC screening in Chinese outpatients.